Investigating Active-learning-based Training Data Selection for Speech Spoofing Countermeasure

TL;DR

This paper explores active learning strategies to improve the generalization of speech spoofing countermeasures, demonstrating that selective data training enhances performance with less data.

Contribution

It investigates active learning methods for training spoofing countermeasures, showing improved generalization and data efficiency over traditional training approaches.

Findings

Active learning improves CM generalization on unseen data.

AL-based CMs perform comparably to using full data pools with less data.

Diverse data inclusion and cautious selection are key for effective AL.

Abstract

Training a spoofing countermeasure (CM) that generalizes to various unseen data is desired but challenging. While methods such as data augmentation and self-supervised learning are applicable, the imperfect CM performance on diverse test sets still calls for additional strategies. This study took the initiative and investigated CM training using active learning (AL), a framework that iteratively selects useful data from a large pool set and fine-tunes the CM. This study compared a few methods to measure the data usefulness and the impact of using different pool sets collected from various sources. The results showed that the AL-based CMs achieved better generalization than our strong baseline on multiple test tests. Furthermore, compared with a top-line CM that simply used the whole data pool set for training, the AL-based CMs achieved similar performance using less training data. Although no single best configuration was found for AL, the rule of thumb is to include diverse spoof and bona fide data in the pool set and to avoid any AL data selection method that selects the data that the CM feels confident in.